Disparities in the Availability of Cancer Clinical Trials: A Multi-level Analysis Participation of minority populations in cancer clinical trials is vital to ensure generalizability of the results generated, facilitate discovery of novel therapies and therapeutic responses that are particularly relevant to traditionally underrepresented populations, and ensure equitable access to new and promising treatments. Yet more than twenty years after Congress mandated that the NIH ensure `sufficient and appropriate' participation of minorities in clinical research, the participation rate of many minority populations remains substantially lower than the composition of the overall US population. Disparities in access to cancer clinical trials are a chief barrier to reducing disparities in trial participation. Thus it is critical to understand the role that geographic variation plays in driving disparities in access to cancer clinical trials if we are to design and target effective interventions to improve minority participation in such trials. Prior studies have had limited success in teasing apart the complex mix of site- and health care system-level factors that influence the availability of different types of cancer clinical trials for minority populations for two chief reasons. First, we currently lack a comprehensive database of sufficiently detailed trial-level characteristics or information about the locations where trials are launched because of the difficulty in synthesizing such information from multiple sources using traditional record linkage methods. As such, prior studies evaluating disparities in the availability of cancer clinical trials have been importantly limited in the range of factors examined and the number and type of trials studied. Second, cancer clinical trials are often launched at multiple sites that operate within larger healthcare systems, and individual recruiting sites often participate in multiple trials, which pose methodological challenges to disentangling the associations between various trial-, site-, and health care system-level factors on cancer clinical trial availability. We propose to address these challenges by integrating information from multiple sources using advanced record linkage methods and applying sophisticated modeling techniques to analyze the resulting multilevel data. The proposed study will improve understanding of how the availability of cancer clinical trials varies across geographic settings and the clinical trials portfolio, which could help target interventions and policies to address disparities in access to the areas most in need. This project will also lay the scientific foundation for a planned future study to predict enrollment of patient populations traditionally underrepresented in clinical research in publicly funded cancer clinical trials.